Bottlehead Forum
Bottlehead Kits => Crack => Topic started by: bigfatpaulie on July 11, 2018, 02:24:18 PM
-
Hello all,
I built my new Crack (this is my second one and 3rd BH amp) and it worked perfectly. Now I added in the Speedball - all the voltage checks passed, all the LED's light up but there is no sound on the right channel.
I've checked continuity on all the paths. I've flowed the solder on every connection on the SB boards and on the joints that I work on tonight. I've switched L/R RCA cables, the side issue stays on the same side. I've tested the headphones on a different amp and they work.
Added: I tested every transistor and diode.
Added: I've checked every resistor.
Added: I've rolled both tubes.
I should also not that the left channel sounds great.
Any thoughts? Could it be a DOA part? What am I missing?
Thanks for any advice.
-Paul
-
Most likely a bad solder joint got jostled when you were changing out parts for the Speedball. Pictures of your build could be helpful, or just reflow every joint. If the voltages are correct, then the circuit is working, and therefore the parts cannot be "DOA."
-
Yeah, I suspect so but this isn't my first build (I've also done three ANK kits, including a Level 5 DAC) and I've never been so stumped. I've even tested the continuty from each post to the next one that the wire is connected too (IE, does the wire actually make a connection).
I've attached some photos. Please let me know if there are others you would like to see. I apologise for the mess, but at this point nearly everything has been re-flowed or re-solderd 3 times...
-
The lock washers are missing from the standoffs — this won't cause the current problem, but it is a missed step.
Since the circuit is working, but not passing signal, the question is where is the signal getting lost. Several places that are somewhat common spots for this kind of thing are not shown in your images - the RCA jacks, the potentiometer (I see you've got an eBay special there, those are not always as reliable as the stock pot), and the headphone jack. Also the grid connections for the right channel: A7 is one of your sketchier looking joints and B1 is hard to see. A6 has too much solder on it, possibly concealing a joint that is not mechanically sound.
-
Thanks.
Thanks for the input.
As for the lock washers the manual says:
Insert a 4-40x¼" screw from the top side of the
chassis through the hole to the left of the 9-pin
socket. Slip a #6 lockwasher over the screw and
secure into a #4x1" nylon standoff.
For whatever reason I took that mean the washer went on the OTHER side of main plate.
As for the RCA jacks there is continuity from them to the pot, resistance checks out on the pot and continuity from the end of the circuit to the headphone jack... As well, keep in mind this did work before the Speedball and those areas are touched with the installation of it.
You're logic (Since the circuit is working, but not passing signal) however, is totally sound. I will go over A7, B1 and A6 and report back.
-
Okay, I cleaned them up a bit and nothing. What is of note, however, is that if I unplug the left RCA cable and turn the volume way up on my source I can ever so slightly hear it on my right side (left is silent).
-
Also meant to mention, most of your joints have too much solder already. When touching up, just hold the iron in place so the solder melts and flows evenly. No need to add more solder
That attenuator doesn't have lugs on it?!? That's just asking for poor joints. (The out end on the right channel is quite blobby...) If you have a pair of sacrificial headphones and a clip lead, you could try jumpering from the RCA hot to the red wire at the output of the attenuator. Note that this will be very loud - a source with a lowerable output is advised and definitely don't use the good headphones.
Also the HP jack looks a little under heated as well.
-
Thank you so much for talking all the time to help me. I really appreciate it.
RE; a lot fo solder. Yeah... Like I said, I've reflowed, and in come cases added solder so there is a lot. The amp is a bit of a mess now.
Okay, I completely bypassed the volume pot and RCA plugs - right from the inputs on the back of the RCA jacks right to A2 and A7 and the black to the ground. Same. Bloody. Issue.
I swear, when I get this I will be screaming from the roof tops.
As for the volume pot being a piece; yeah. I didn't know until it showed up. Hey, at least it was cheap. I don't use it so I could remove it entirely and just put in some resistors.
-
Well if bypasssing it doesn't solve it, then that rules out its culpability.
I'm leaving the office for the night. You said your voltages were good, but could you please post them in case something stands out?
-
Of course. No problem:
OA : 77.4
IA : 196
B-A/B : 0.01MV
IB : 195.8
OB :70.4
OA : 104.3
OB : 123.7
G : 0.03 MV
B+ : 195.6
-
FYI on HF a member asked me to test the tube voltages and they were:
A3: 1.53
A8: 1.53
B3: 137.6
B6: 103.7
Which prompted:
"somewhere in the circuit, from B3 to ground is not connected. so the tube is exhibiting the plate voltage (or near plate voltage) on the cathode, because its not conducting (fully). I would check the transistor's ground connection. R1 to ground of the constant current source board, and test the transistors (you don't have to pull them out in this circuit). Since its brand new, its not unheard of a 2n2222 to be bad brand new, thats why they are cheap (under a $1 each) and the tip50 is about half of that price."
Is that possible? How would I test that?
-
Permission is not granted to post any portion of our manuals on any forums.
That guy is very impressed with himself. I have asked the kit's designer to take a look at this thread and he should be helping you shortly.
-
To bring down your B3/big board OB voltage, consider resoldering the joints on the TIP50 by OB. The center leg can be difficult to properly solder as it is connected to the heatsink. This iffy voltage could be causing the signal to drop.
To trace the signal, start by downloading a 60Hz tone onto your phone, then play that into the amp. With the amp off, let us know the AC voltage you see with the volume control all the way up at A2 and A7. Once you report those voltages, we can move along to the other parts of the amp.
The advice you got on HeadFi about the 2N2222 and TIP50 being cheap and unreliable is nonsense.
-
Yeah, it seemed he was trying to help but he couldn't help bashing every part. When someone is that upset about something they lose perspective... To each their own I suppose.
Thank you both for your help.
Okay, I have re-flowed the 6 solder joints on the 2 TIP50's. I help the top of the iron on the underside of the board and used some very fine solder at on the top side letting is flow down though the board.
No change.
As for the 60Hz test I have a few questions. So I did it (https://www.youtube.com/watch?v=3Mjc5JJcu5I from my MavBook) and right at the RCA input my multi-meter was read a fluctuating number between 0 and 0.09mV. So I unplugged the RCA cable and measured it directly at the plug and I am getting the same result. Is that correct or am I measuring something wrong? What number (ish) should I be getting? My 0 - 0.09 doesn't seem right.
-
What does the left side measure? Is the volume cranked on the laptop?
-
Yes. YouTube is maxed and laptop volume is maxed. If I plug in headphones I can clearly hear it.
Both sides of the RCA cable measure the same. Here's a video of what I am doing:
htt p://wri stshot wed.com/per sonal/IMG_12 46.MOV
(remove the spaces, as the forum isn't posting it correctly)
-
You'll want the volume all the way up on your source (phone) when you're doing that test. You should see about 0.3V of AC between the center pin and shell of the RCA plug.
Did your voltages change with that resoldering?
-
AC! Okay... I was thinking DC. I am getting 0.473 at the cable then. Let me test.
RCA R: 0.473
RCA L: 0.473
A2: 0.473
A7: 0.473
-
OK, that clears your input wiring. Next you can run the amp and check AC voltage between ground and 1, then ground and 5. Your meter may not return a reading since there's DC present, but some will be able to resolve the AC measurement there.
Did your OB voltage on the big board drop after resoldering?
-
Okay. Thank you.
1: 9.12 (AC) 77.7 (dc)
5: 8.94 (AC) 70.8 (dc)
OB: 167.7v (???)
-
Yeah, half the big PC board just isn't conducting. Can you post photos of the top and bottom of the board? 99% of the time, this is just a cold solder joint, I suppose 1% of the time the wire from the "O" terminal is broken or not landing on the correct terminal.
-
Thanks! Hopefully we (you) are getting closer :)
I really appreciate all your effort and patience on this.
PS. Sorry for the crummy photos, it is hard to get a photo of it attached.
PPS. I also tested the continuity of the connections to board again.
PPPS. I've since re-flowed every point on the big board. No change.
EDIT: Here's something interesting. If I put my headphones on with the amp powered up (but nothing playing) and I measure the voltages at OA and OB I hear static ONLY in my left channel, never in the right.
-
https://www.youtube.com/watch?v=VgcPxdnjwt4
You can go longer than 2 seconds, and especially on the center of the TIP-50, probably should.
-
Please keep in mind I've re-flowed ALL these solders AT LEAST three times. At this point, no, they don't look perfect...
-
I know they do not look perfect. That is why I posted a video showing what good solder joints look like and clearly showing the process to get good looking solder joints.
-
I get that. It's really the best I can get them to look at this point, however. As I said, this didn't start out looking this way... Keep in mind this isn't my first Crack/SB. I've also built a Mainline and several ANK kits including the level 5 (https://www.head-fi.org/threads/audio-note-kit-dac-5-1-signature-nos-r-2r.793641/).
I may not have the knowledge to diagnose, but I have a good amount of soldering in my day and I have been working with electronics for 15 years. I'm certainly no professional, but I really suspect there is something else going on in this case.
Is there any possibility, at all, that one of the transistors isn't functioning correctly?
-
At least one of the center legs of a TIP50 transistor does not have a solid looking joint on it. Try putting your iron on the pad/center lead, then count to 20, then pull it off.
-PB
-
I usually desolder my joints with a desoldering pump if I noticed anything that didn't look good. When I tried reflowing my joints sometimes they looked the same or worse so I always made a habit of doing them over. I just never knew you could reflow them by reheating it that's why I always did it like that. Not sure if it will help in your case, but I'd give it a try on the Speedball board if nothing else works.
-
Thanks, guys.
Okay, I completely removed the big board from the amp and did the 20 second soldering iron 'hold' on every joint, not just the ones on the TIP50. No change.
Here is an odd symptom. I don't know if the amp was doing this before or just started as I didn't try it prior to today.
If there is music playing at the source and I power up the amp with headphones on, I hear the music FIRST come in on the right side then it quickly shifts to the left fully as the amp powers up, leaving the right side dead. The best way I can articulate it is if someone was on the right of my then quickly moved to my left.
I don't know if that helps but that's what I am experiencing.
-
It's a really, really bad idea to listen to an amp that doesn't pass its voltage tests. You may end up being in the market for new headphones.
Can you post new photos of the top and bottom side of the big PC board?
-
It's a sacrificial pair of old HD433's - I'm okay if they blow up.
I, again, will point out that each time I am testing the continuity from a different place on the board to the other side of the component to ensure continuity (IE, testing if the actually solder joint is conducting) AND measuring the resistance. If it isn't 0 ohms, I redo it.
However they look, I would honestly bet my right hand that the solders are all working...
So I'll ask again, is there, however small, a possibility that a transistor isn't functioning correctly?
PS, sorry but the bad photos
-
Could you please remove the board and photograph it in focus and without backlight?
While the board is out, you could re-install the 3K resistors and confirm that the amplifier passes signal without the large Speedball board.
Also are you checking voltages after resoldering? You have not specified how you are determining that there is no change, and the only unchanged symptom you have listed is the dead channel.
-
I, sadly, don't have the resistors anymore. Like an idiot, I thought this was going to a simple upgrade...
When I say 'no change' I mean the right channel is still dead.
Wish I could get an answer to my question - is it at all possible that one of the parts has failed?
-
Could I substitute film resistors to test?
-
I still see cold solders on those TIP50 transistors.
Are you, by chance, using lead free solder? What kind of iron are you using? Heat and lead will be your friends for getting a good joint.
-
Hello,
It's a Solomon SL-10 soldering station and it is "Rosin core Solder for PC wok". It does contain lead.
I realize that solders look a little gray, but it is the photo and reflection. In person they are all shiny.
Are there any other test/measures I can do? Or should I just start buying replacement transistors?
BIG BOARD:
OA: 103.1
OB: 148.7
G: 0
B+: 196.0
SMALL BOARD:
OA: 77.7
IA: 196.6
B-A/B: 0
IB: 196.1
OB: 70.6
-
It's not the sheen or lack thereof, those joints just need a little additional solder.
A shorted transistor will measure very low impedance between a pair of legs. In the absence of such a measurement, the transistor in question is likely working properly.
-
I don't mean to be rude, but this is a little frustrating... You said cold solder joints and in the manual it says: A cold joint will usually appear as a convex blob and be dull or even crystallized in appearance."
Now you are saying it isn't enough solder, which is different.
I'm confident that 'not enough solder' isn't the issue. See the image - I've ONLY soldered from the underside of the board which means this is solder that has flowed down. There is A LOT of solder... Are there any other voltage test/results I can give you? Is there another possibly aside a bad solder joint? Like anything, however remote, at all possible?
Please.
-
We all sometimes get frustrated when we are sure that we have done what has been suggested and things still don't work. I'm looking at that middle pin of the TIP50, and I have to say the first thing that comes to me is that I would reheat it until I saw that blob melt and create a fillet with the solder pad. Generally the best approach when reflowing a joint like that is to have a solder spool handy so you can add just a little bit more solder of it looks like it is necessary to get a fillet to the pad all the way around the lead. We're not trying to frustrate you, we are trying to help.
-
That's the thing, please don't get me wrong: I really appreciate you guys trying.
But, truly, I can't help but feel after re-flowing the solder three times, then de-soldering everything, then re-soldering then re-flowing two more times - doing all that and each time I have the exact same issue... Not to mention things like the TIP50's have been reflowed probably another 4 - 5 times and de/re-soldered additionally.
And does the shift from the right side to the left side not strike either of you are odd? I mean, have you ever experience that, and if so was it a bad solder joint?
Also - can I use film resistors and remove the big board? I realize they will get really hot, but I just want to see if I can get both channels working, even if for just a moment.
For your review before I reinstall the board yet again, how do these solder joints look? Please note that I have only added solder to the bottom side of the board, everything on the top side had flowed though.
-
Those solder joints look a lot better, let us know what voltages you get with the board in that condition.
Using a pair of film resistors in place of the large Speedball board is possible, but you'd need a pair of 3K/3W parts. These will get insanely hot, but they shouldn't self destruct too quickly. Anything under this power rating is going to smolder rather quickly.
-
Thanks. Okay, no dice. So I desoldered the TIP50's again and re-soldered. Again, no dice.
So I took some of the kit wire and bridged from each of the three pins to their locations on the PCB and HAZZAH! Sound! I don't know if the wire is helping the connection, or if there is a brake on the board or whatever, but I am getting sound now.
That said, the sound on the right side is quite a bit quieter than it is on the left... IE, the challenges aren't balanced now.
Is there anything I can do to fix this?
-
Post your new voltage measurements.
-
OA: 102.3
OB: 112.5
G: 0
B+: 189.8
-
Now test to see if the volume balance is due to the volume pot. To do this, plug your phone into the Crack with the phone level very low and the Crack's volume pot nice and high.
-
No change. I tried with the volume on the Crack all the way and stepped down to see if there was a change in balance and there isn't.
I even jumped from the RCA in to past the volume pot (IE, eliminated the volume pot from the chain) to make sure.
-
No suggestions at all?
We've come so far! :)
-
Have you thought of sending it to BH for resolution?
Alex
-
With the amp on and the 60Hz tone playing at the input, what is the AC voltage at 6 & 10?
-
Thank you.
6: 0.410 VAC
10: 1.378 VAC
-
With the amp on and the 60Hz tone playing at the input, what is the AC voltage at 1 & 5?
-
1: 20.86v
5: 20.26v
-
The only suggestion that makes sense with these measurements would be to try another 6080. If that doesn't solve your issue, we are approaching the end of what can be done without having the amplifier on the bench. The repair service (https://bottlehead.com/?product=repair-service) exists for these kinds of cases.
-
Okay - I swapped out the 6080 for a NOS 5998 tested and sold by Brent Jesse (ie, it's perfect) and there still is a balance issue, and sadly, the voltages test the same...
Part of me feels good that it isn't an obvious issue.
Part of me is really disappointed as clearly is something going on here.
To pay $165 to diagnose a $199 doesn't make sense: I would just as soon throw it out and buy a new kit, frankly. Don't get me wrong, I'm not knocking the service, I get that it's a flat rate on a $199 and on a $1999 amp. I also really appreciate all the time and effort everyone has put into this.
Out of interest, what's all between the 1 & 6 and 5 & 10? The big board and the 6080: is there anything else? I am considering rebuilding the big board and replacing all the parts with new as it's really only about $5 worth of stuff on it. I just want to know if there anything else I should be swapping out.
-
It's possible (but I can't say certain, without having the amp on the bench) that the big board might have some fault on it and it may be a damaged semiconductor. If you found a damaged trace (where you laid the piece of wire over it) that should be considered an omen that trying to rework a board with new parts is potentially asking for trouble. If you decide that replacing the board will fix the amp you might be playing it more safe to start with a fresh PC board, and we can sell you the board and parts. Before you go there double check that the output cap on the bad side didn't get brushed by a hot iron and burned. You can contact Eileen at queen at bottlehead dot com to order the parts. We are pretty much out of other ideas without being able to put the amp on our bench and go through it.